Multimedia Transmission for Emergency Services in VANETs Muhammad A. Javed, Jamil Y. Khan, and Duy T. Ngo School of Electrical Engineering and Computer Science The University of Newcastle, Callaghan, NSW 2308, Australia Email:
[email protected],
[email protected],
[email protected]
Abstract—Emergency warning applications in a vehicular ad hoc network (VANET) requires successful dissemination of warning notification within a geographical area spanning multiple hops. Multimedia transmission provides an accurate overview of the emergency event in the form of an image, an audio or a video. However, the reliable transmission of emergency multimedia messages using multi-hop broadcast techniques suffer from broadcast storm and severe interference from the existing periodic single-hop safety messages. In this paper, we propose an efficient time-slotted multi-hop broadcast protocol for emergency multimedia transmission. To avoid interference with the safety messages already in place, we propose to allocate separate multihop time slots for the multimedia messages, and devise a signaling mechanism that ensures a reliable delivery of these multimedia messages. The proposed mechanism can also handle the scenario of lost acknowledgment (ACK) and thus effectively reduce the unnecessary retransmissions of the multimedia messages. Simulation results confirm that the developed protocol substantially outperforms existing schemes in terms of both the number of required multi-hop transmissions and the dissemination delay, while maintaining a high reception rate and a low end-to-end delay for the single-hop safety messages.
I. I NTRODUCTION Vehicular Ad hoc Network (VANET) is a key component of the future intelligent transportation systems that support safety, traffic management, and user infotainment applications [1]. In a VANET, multimedia communication could be employed to provide vehicles with information about ongoing road emergency events, which include accidents, buildings on fire, floods, road closure, car breakdown, etc. The emergency message may contain images of an accident, short video of a building on fire or an audio message that gives information about flood on the highway [2], [3]. Compared to a simple text message, such multimedia messages provide more precise and informative details of the emergency situation, enabling other vehicles on the road to better decide on how to react to the event. An improved awareness of the emergency can also help the rescue teams to plan in advance, and thus more effectively handle the emergency situation. Multimedia communications in a service network such as VANET is different from conventional multimedia communications in that the information is transmitted in a multicast or broadcast manner and the communication nodes operate in an ad hoc manner. Additionally, as the emergency multimedia message has to be sent to locations potentially out of a vehicle’s transmission range, the communication takes place over multiple hops. Ranging from several hundred kbytes to
a few Mbytes, a multimedia message is typically decomposed into many small fragments to be reliably transmitted to the destination vehicles. The destination vehicles then combine all the fragments to reconstruct the original multimedia message. The total transmission delay of a multimedia message is limited to a few seconds to ensure a timely notification [4]. The efficient dissemination of emergency multimedia messages over multiple hops faces several major challenges. The emergency multimedia messages may be severely interfered by the frequently-generated safety messages whose packet transmission requirements are generally stringent [5]. In relayassisted multi-hop transmissions, the the excessive redundant information introduced by broadcast storm may also worsen the network congestion situation. The interfering situation may be further complicated by the hidden nodes where packet collisions cannot be detected [6]. To efficiently disseminate multi-hop warning messages in VANETs, Urban Multi-hop Broadcast (UMB) [7] protocol partitions the road into small segments and uses a handshaking mechanism for relay selection. In timer and probability based protocols, the furthest vehicle in the range of the sender forwards the message due to assignment of a shorter wait time or probability of transmission [8]. A similar timer based approach is used by the Contention-Based Dissemination (CBD) protocol for relay node selection [9]. In [10], the proposed Distributed Vehicular Broadcast (DV-CAST) scheme employs the slotted 1-persistence approach to suppress broadcast storm and also adopts the store-carry forward mechanism for disconnected networks. Few other multi-hop techniques proposed in literature can be found in [11]–[14]. In this paper, we propose an efficient time-slotted protocol for emergency multimedia warning transmissions in a VANET. We divide the emergency multimedia message into a number of fragments and allocate several multi-hop time slots for the transmission of each fragment. Here, a multimedia message sender reserves a time slot by first transmitting a long-range CLEAR packet to inform all the potential hidden nodes of an incoming transmission. The multimedia message can then be sent without being interfered by the single-hop safety messages. Note that should there be no multimedia messages, we allow the time slot to be used by the safety messages, thus fully utilizing the network resources. To further increase the reliability of the multimedia transmission, we include a short acknowledgment (ACK) packet as part of the multi-hop time slot. We also propose a mechanism
to reduce the unnecessary retransmissions in the case of the ACK packet being lost. Compared with the existing multihop protocols namely Contention Based Dissemination (CBD) [9] and Distributed Vehicular Broadcast (DV-CAST) [10], our multi-hop broadcast protocol design offers a significant reduction in the number of number of multimedia message transmissions and the associated dissemination time, while providing a high reception rate and a low end-to-end delay to the periodic safety messages. II. P ROPOSED T IME -S LOTTED M ULTI - HOP B ROADCAST P ROTOCOL FOR M ULTIMEDIA M ESSAGE D ISSEMINATION
Fig. 1. Structure of a time slot in the proposed multi-hop broadcast protocol
A. Multi-hop Time-slot Reservation Mechanism
(4)
(CSMA/CA) mechanism. In this BACKOFF phase, safety message transmissions are also suspended. The vehicle whose backoff timer expires first will then transmit a CLEAR packet at a transmission range Rt . The purpose of the CLEAR packet is to reserve the rest of the time slot for the multimedia packet, and also to inform the vehicles in the range Rt of the upcoming multimedia fragment transmission. The CLEAR packet also contains the additional information of Tf and N that informs all vehicles about the number of slots within Tf and start time of each multimedia slot. Upon receiving the CLEAR packet, other vehicles suppress their transmissions until the end of the time slot. As such, all the vehicles in the range Rt who intend to send periodic safety messages during the reserved multimedia time slot will not interfere with the multimedia fragment transmission. Vehicles who do not receive a CLEAR packet may continue with its safety message transmission as soon as the backoff time is over. This arrangement allows the full utilization of the multihop time slot in the absence of an emergency multimedia fragment. After the CLEAR phase, the vehicle who is reserved with the multimedia time slot is allowed to send the multimedia fragment (i.e., DATA) at a transmission range of Rt /2. Since the CLEAR packet is transmitted over a range of Rt , all hidden nodes located within two transmission hops from the multimedia-fragment sender are made aware of the upcoming multimedia transmission. Essentially, the interference from any hidden nodes during the transmission of such a message is eliminated. In the CONTENTION phase, each vehicle who has received the multimedia fragment computes a contention time Tcnt . We propose that the value of Tcnt is inversely proportional to the distance between the sender of the multimedia fragment and the receiver: (Dr − Ds ) Tcnt = · Ts , (5) Dc
The proposed multi-hop time slot starts with a random BACKOFF phase, during which vehicles with queued multimedia packets wait for a random time. It is noteworthy that the safety messages are uniformly generated within a synchronization interval (SI) of 100ms, and that they employ the carrier-sense multiple access with collision avoidance
where Ds and Dr is the position of sender and receiver vehicles, Dc is a constant which ensures that all vehicles Dc distance apart away have a different contention time, and Ts is the slot time. In the ACK phase, the vehicle with the shortest contention time Tcnt will transmit an ACK packet. It is worth recalling that this vehicle is responsible for relaying the multimedia
Due to the large size of a multimedia message, we decompose the message into several fragments, each of which is to be reliably transmitted to the receiver. Upon combining all the received fragments, the receiver can recover the original message. Let Sm be the size of a multimedia message (in bytes) and Sf the size of a single fragment (in bytes). The number of packets required to completely send the multimedia message can be written as: P =
Sm . Sf
(1)
Let Tm be the time needed for the entire multimedia message to be transmitted. This implies that a single fragment must be transmitted within: Tm Tf = . (2) P For emergency multimedia message transmissions, we propose to divide Tf into N time slots, each of which has the structure depicted in Fig. 1. Let Da be the distance between the emergency warning vehicle and the end point of the geographical area of interest. Due to multi-path fading, we assume the reliable transmission range as one-half of the actual transmission range Rt . The number of time slots required for a single multimedia fragment transmission can be calculated as: Da Na = , (3) Rt /2 which is exactly the number of hops required to disseminate the multimedia message within area Da . To make sure that the fragment is transmitted to all the vehicles in the geographical area within the Tf duration, we propose the use of an additional number of guard time slots Ng . Altogether, the number of time slots required for a single fragment transmission is: N = Na + Ng =
2Da + Ng . Rt
TABLE I S IMULATION PARAMETERS Parameter Highway Road
Fig. 2. Lost ACK scenario.
fragment in the next time slot. After receiving the ACK of the multimedia fragment from the contention winning vehicle in the current time slot, all other vehicles (i.e., the potential forwarders) delete the corresponding multimedia fragments in their queues. No other actions are required from these vehicles. Note that such a message cancellation policy only applies when a ACK is received from a vehicle located further away in the message propagation direction. This ensures the progress of the multimedia fragment in the propagation direction. It might also happen that the vehicle who has forwarded the multimedia fragment does not receive an ACK within the current time slot. In such a case, this vehicle will resend the same multimedia fragment in one of the next time slots that are reserved for multimedia message dissemination.
Vehicle
Safety message
Multi-hop BACKOFF Contention Window
It is possible that the ACK packet sent by the vehicle responsible for relaying the warning message might get lost during transmission. In the absence of an ACK, other vehicles would have to resend the warning message unnecessarily. While exact the number of redundant retransmissions depends on how many vehicles have received the ACK packet, it can be substantial in poor channel conditions. In this paper, we propose the following arrangements to handle the ACK loss situation and thereby improving the efficiency of channel utilization. Illustrated in Fig. 2 is a warning message transmission scenario, where all 4 vehicles are potential relay vehicles. Suppose that the warning message sent by the vehicle in emergency is received by Vehicles 1 − 4. As the furthest from emergency vehicle, Vehicle 4 has its contention timer expired first [see (5)] and thus sends an ACK4 packet. Suppose that this ACK4 packet is not received by all other vehicles, e.g., due to severe interference or deep channel fading; only Vehicle 3 receives this ACK4 packet. Without any ACK4 reception, Vehicle 2 believes that it is responsible for relaying the warning message and thus sends an ACK2 packet after the CONTENTION phase. In the next time slot, both Vehicles 2 and 4 will have the warning messages (originally from emergency vehicle) in their queues, to be relayed to further nodes. If Vehicle 4 wins in the CONTENTION phase of this time slot, it will broadcast the warning message. In this case, Vehicle 2 will receive a duplicate data packet from a vehicle located further away in the direction of message propagation (i.e., Vehicle 4). Knowing that the warning message has progressed further, Vehicle 2 can just delete the warning message in its queue, and take no further action. On the other hand, if Vehicle 2 wins in the
Road length No. lanes Medium density High density Speed Size Data rate Transmission range Generation freq. Size Slot time TS
2km 6 (3 per direction) 120 (vehicles/km) 180, 240 (vehicles/km) 60, 45, 30 (km/h) 500 bytes 6Mbps 300m 10Hz 7 13µs
CLEAR packet
Size Transmission range
24 bytes 1, 000m
DATA message
Size Transmission range
528 bytes 500m
ACK packet
Size Transmission range
38 bytes 500m
Multimedia message parameters
B. Handling the Lost ACK Scenarios
Value
Tm Image resolution Sm P Tf Na Ng Fading model
20s JPEG (340 × 160) 50kbytes 103 0.194s 8 4 Nakagami-m (m = 3)
Reception Rxth
−91dBm
Background noise
−99dBm
Simulation time
300s
CONTENTION phase, it will broadcast this warning message. In this case, Vehicle 4 will receive a duplicate data packet from a vehicle at a lesser distance in the direction of message propagation (i.e., Vehicle 2). Vehicle 4 will then reply with an ACK4 packet to notify Vehicle 2 that the warning message has progressed further, and no other action is required from Vehicle 2. In either scenario, the unnecessary retransmissions of the warning message are avoided. III. P ERFORMANCE E VALUATION We evaluate the performance of our proposed Emergency multimedia time-slotted protocol (EMTS) protocol using OPNET Modeler 16.0. Specifically, we consider a highway scenario with a road length of 2 km and assume there are 3 lanes in each opposite direction. At medium vehicle density (120 vehicles/km) we assume an exponentially-distributed intervehicle spacing, whereas at high vehicle densities (180 and 240 vehicles/km) we assume normally-distributed inter-vehicle spacing [15]. To represent a medium fading intensity, we assume Nakagami-m fading with m = 3. We assume that every vehicle generates periodic safety messages with a transmission range of 300m and at 10-Hz rate. We place an emergency warning vehicle that generates and transmits a multimedia image message within the 2km
road section. Assume that the multimedia image is in JPEG format with a resolution of 320 × 160 pixels. If 1 byte per pixel is required to store the red, blue and green (RGB) light values, the image size is Sm = 50 Kbytes. Let the fragment size be Sf = 500 bytes, Tm be 20s and the number of guard time slots be Ng = 0.5Na . The practical parameters used in our simulations are listed in Table I. Referring to the time-slot structure of our proposed design in Fig. 1, the multi-hop time slot size Tts is computed as: Tts = Tb + Tc + Td + Tcnt + Ta ,
(6)
where Tb is the BACKOFF time, Tc is the CLEAR packet transmission time, Td is the DATA (multimedia message) transmission time, Tcnt is the CONTENTION time, and Ta is the ACK packet transmission time. Here, the backoff is selected as a random integer from a contention window consisting of 7 slots, each of which is 13µs in duration. The contention time Tcnt is determined by (5), where the Dc is taken as 10m. The size of a CLEAR packet is 24 bytes (8 bytes of physical overhead, 8 bytes of Tf and 8 bytes of N ), and it has a transmission range of 1, 000m. The DATA message is 528 bytes long (including 500 bytes of multimedia fragment, and 28 bytes of MAC overhead), and its transmission range is 500m. ACK is a short packet of 38 bytes (including 30 bytes of ACK information, and 8 bytes of MAC overhead) and its transmission range is 500m. The multi-hop time slot size Tts comes out to be 1.67ms. We compare our proposed EMTS transmission design with two existing protocols, namely, CBD [9] and DV-CAST [10]. In our simulation study, the transmission range, the contention area and the maximum contention time for the CBD protocol are set as 500m, 300m and 50ms, respectively [9]. On the other hand, the transmission range, the number of time slots and the maximum wait time for the DV-CAST protocol are taken as 500m, 5 and 5ms, respectively [10]. In Fig. 3, we show the reception rate of multimedia fragments within the 2km road section. As evident from the results, our proposed solution guarantees an almost 100% delivery rate in every vehicle density scenario considered. This is a noticeable enhancement in light of the 91 − 94% reception rates provided by the DV-CAST and the CBD protocols. The performance of these protocols is degraded due to interference between the existing safety messages and the multimedia messages resulting in the loss of multimedia fragments. Fig. 4 shows the fragment delay, i.e., the average time period required to disseminate a single fragment of a multimedia image within the 2km road section. It is clear from the figure that the proposed EMTS protocol has a much lower dissemination time compared to both the DV-CAST and the CBD protocols. In particular, at the density of 120 vehicles/km the EMTS fragment delay is 10ms shorter than the DV-CAST and the CBD counterparts. As the number of vehicles per km increases, the fragment delay incurred by the DV-CAST and the CBD schemes dramatically grows due to the higher interference from the periodic safety messages. However, the effect of interference is limited in the case of our EMTS proposal since we employ separate time slots to send the
Fig. 3. Reception rate of emergency multimedia fragments within a distance of 2km.
Fig. 4. Time required to disseminate a single fragment within a distance of 2km.
multimedia message. From Fig. 4, the resulting EMTS fragment delay only slightly increases, remaining below 115ms for all the vehicle densities under consideration. Finally, the most pronounced advantage is observed at the vehicle density of 240 vehicles/km, where the fragment delay is dropped by 30 − 42ms by our proposed EMTS solution. We display in Fig. 5 the average number of transmissions required to successfully disseminate a single fragment of an image message to the vehicles within 2km road section. In comparison to the DV-CAST and the CBD protocols, our proposed EMTS scheme significantly reduces the number of required transmissions. Particularly, at the vehicle density of 240 vehicles/km, only 7 transmissions are needed for dissemination of a single fragment. This figure represents a mere 7 − 10% of the total number of transmissions required by the existing approaches. Such a remarkable gain is a direct result of employing separate time slots for multimedia message transmission resulting in interference avoidance from the periodic safety messages. Moreover, the CLEAR packet transmission results in alleviation of hidden node collisions and the arrangement to handle lost ACK scenario reduces the number of transmissions. We also examine the effects of our proposed multimedia message transmission protocol on the existing periodic singlehop safety messages. It is apparent from Fig. 6 that while the reception rate of the safety messages is degraded in presence of multimedia message, the effect is mild in our EMTS protocol.
Fig. 5. Average number of multi-hop transmissions required to successfully disseminate a single multimedia fragment within a distance of 2km.
Fig. 7. End-to-end delay of the safety messages within a distance of 100m.
R EFERENCES
Fig. 6. Packet success rate of the safety messages within a distance of 100m.
At the density of 240 vehicles/km, the loss of safety messages caused by the proposed EMTS multimedia transmissions is simply 6% away from that in the case of no multimedia messages. This is a clear improvement from the 11% and 10% losses resulting from the DV-CAST and the CBD protocols. A similar trend can also be observed from Fig. 7, albeit in terms of the end-to-end delay of safety messages. Here, our EMTS solution offers a 10 − 20% reduction in the end-to-end delay, while following quite closely with the delay in the case of no multimedia transmissions. IV. C ONCLUSIONS In this paper, we have presented a multi-hop broadcast protocol to efficiently disseminate emergency multimedia messages in a VANET. To reduce the mutual interference with the existing periodic safety messages, we have proposed a timeslotted structure for the multi-hop transmissions. Specifically, we have assigned separate time slots for the multimedia messages, and proposed the use of a CLEAR packet before the actual data transmission to eliminate all the hidden nodes in range. To avoid the unnecessary retransmissions of the warning messages, we have established a mechanism that effectively handles the case of ACK packet losses. Simulation results with realistic parameters have verified the clear advantages of our proposed scheme over existing solutions in several key performance criteria.
[1] Y. Toor, P. Muhlethaler, and A. Laouiti, “Vehicle ad hoc networks: applications and related technical issues,” IEEE Communications Surveys Tutorials, vol. 10, no. 3, pp. 74–88, Sep. 2008. [2] Z. Yang, M. Li, and W. Lou, “CodePlay: Live multimedia streaming in VANETs using symbol-level network coding,” IEEE Transactions on Wireless Communications, vol. 11, no. 8, pp. 3006–3013, Aug. 2012. [3] J.-S. Park, U. Lee, S. Y. Oh, M. Gerla, and D. S. Lun, “Emergency related video streaming in VANET using network coding,” in Proc. Intl. Workshop on Vehicular Ad Hoc Networks, Sep. 2006, pp. 102–103. [4] F. Naeimipoor, C. Rezende, and A. Boukerche, “Performance evaluation of video dissemination protocols over vehicular networks,” in Proc. IEEE Intl. Workshop on Performance and Management of Wireless and Mobile Networks, Oct. 2012, pp. 694–701. [5] M. Javed and J. Khan, “Performance analysis of an adaptive rate-range control algorithm for VANET safety applications,” in Proc. IEEE International Conference on Computing, Networking and Communications, Feb. 2014, pp. 1–5. [6] M. Javed, J. Khan, and D. Ngo, “Joint space-division multiple access and adaptive rate control for basic safety messages in VANETs,” in Proc. IEEE Wireless Communication and Networking Conference, 2014, pp. 1–5. ¨ uner, and U. Ozg¨ ¨ uner, “Urban multi-hop [7] G. Korkmaz, E. Ekici, F. Ozg¨ broadcast protocol for inter-vehicle communication systems,” in Proc. ACM Intl. Workshop on Vehicular Adhoc Networks, Oct. 2004, pp. 76– 85. [8] N. Wisitpongphan, O. Tonguz, J. Parikh, P. Mudalige, F. Bai, and V. Sadekar, “Broadcast storm mitigation techniques in vehicular ad hoc networks,” IEEE Wireless Communications, vol. 14, no. 6, pp. 84–94, Dec. 2007. [9] M. Torrent-Moreno, “Inter-vehicle communications: Assessing information dissemination under safety constraints,” in Proc. Conf. on Wireless on Demand Network Systems and Services, Jan. 2007, pp. 59–64. [10] O. Tonguz, N. Wisitpongphan, and F. Bai, “DV-CAST: A distributed vehicular broadcast protocol for vehicular ad hoc networks,” IEEE Wireless Communications, vol. 17, no. 2, pp. 47–57, Apr. 2010. [11] M. Li, K. Zeng, and W. Lou, “Opportunistic broadcast of event-driven warning messages in Vehicular Ad Hoc Networks with lossy links,” Computer Networks, vol. 55, no. 10, pp. 2443–2464, Apr. 2011. [12] J. Sahoo, E.-K. Wu, P. Sahu, and M. Gerla, “Binary-PartitionAssisted MAC-Layer broadcast for emergency message dissemination in VANETs,” IEEE Transactions on Intelligent Transportation Systems, vol. 12, no. 3, pp. 757–770, Sep. 2011. [13] M. Slavik and I. Mahgoub, “Spatial distribution and channel quality adaptive protocol for multihop wireless broadcast routing in VANET,” IEEE Transactions on Mobile Computing, vol. 12, no. 4, pp. 722–734, Apr. 2013. [14] R. S. Schwartz, R. R. R. Barbosa, N. Meratnia, G. Heijenk, and H. Scholten, “A directional data dissemination protocol for vehicular environments,” Computer Communications, vol. 34, no. 17, pp. 2057– 2071, Nov. 2011. [15] K. Abboud and W. Zhuang, “Modeling and analysis for emergency messaging delay in Vehicular Ad hoc Networks,” in Proc. IEEE Global Telecommunications Conf., Dec. 2009, pp. 1–6.
