User Case Analysis for Communication Resilience

One of the defining trends of our century is the rapid urbanisation in both developed and developing worlds. Across the planet, more than 50% of the population is now living in cities, and this is set to rise rapidly over the next decade. Modern cities are partly defined by a high population density and high mobile devices usage (includes mobile vehicles). Therefore, there is an opportunity to achieve multi-hop communications between users. One of the critical challenges global cities face is security from terror attacks. Terrorist attacks generally target dense urban areas to deliver the greatest casualty and a high impact. In the event of such an attack, such as the 9/11 attack in New York City and the 7/7 bombing in London, the mobile networks became overloaded due to the increase phone and date usage. This is typically managed by implementing a class access bar for emergency services, meaning that critical communication can still take place.

96_{Yuan, H., Maple, C. and Ghirardello, K. (2018) August. Dynamic route selection for vehicular store-carry-forward}

networks and misbehaviour vehicles analysis. In 2018 IEEE 88th Vehicular Technology Conference (VTC-Fall) (pp. 1-5). IEEE.

Figure 21. A simulation of V2V emergency communication under terrorist attack

7.3.1. Emergency V2V Communications Under Terrorist Attack

In this section, the use case scenario is that the public access network is under attack, and we lose some access points such as RSU or BS. It should be stated that O2’s vision for C- V2X, would not provide V2V communication via RSUs, but rather directly vehicle-to-vehicle, negating the need of the RSU. In this situation, the communication network is fully loaded with data traffic, and a large set of vehicles and users are seeking alternative ways to relay vital data. Peer assistance V2V communication is a way of allowing vehicles to act as relays for each other97_{. The RSUs of the C-V2X network are then not required for data-}

bearing channels or to serve as a coordinator or facilitator to V2V channels. In Figure 21, it is a use case that shows the communication resilience of V2V

communication in an urban environment under terrorist attack. In this case, two V2V routing algorithms are addressed, the shortest path routing (SPR), which is a greedy path selection algorithm. SPR seeks to minimise the total multi-hop distance or the number of hops in order to improve the multi-hop V2V transmission reliability. In SPR, each V2V knows its location and that of the destination user. Each UE that holds the message will first identify the UEs to which it can reliably transmit and then transmit to the one that is closest to the destination UE.

The other routing algorithm is Interference Avoidance Routing. While algorithms such as SPR can yield reasonable performance and minimise the delay, it may not always yield the best reliability performance. This is because when cross-tier interference between

conventional communication (CC) and V2V transmissions is considered, selecting the shortest path is not always the optimal strategy.

97 _{Yuan, Hu, Weisi Guo, and Siyi Wang (2014) "Emergency route selection for D2D cellular communications}

during an urban terrorist attack." In 2014 IEEE International Conference on Communications Workshops (ICC), pp. 237-242. IEEE

Cross-tier interference is the lowest when the V2V transmissions occur at the RSU

coverage boundary (cell edge). An edge routing path would reduce the V2V interference to CC transmissions in the uplink (UL) band and would reduce the CC interference to V2V transmissions using the downlink (DL) band. The interference avoidance routing (IAR) algorithm tends to migrate along the cell edge in order to trade off a longer route for reduced interference.

7.3.2. V2V Resilience Performance under CC Constraint

One of the key advantages of IAR routing over SPR routing is that it reduces the interference emitted to regular CC UEs. By picking a routing path that travels

predominantly along the traditional coverage edge, it maximises the distance to the majority of CC UEs. The paper now expands the IAR routing to both consider uplink (UL) and downlink (DL) bands.

Figure 22. V2V outage probability for various CC outage constraints

Figure 22 shows the V2V outage probability for various CC outage constraints. The results show that there is an intuitive trade-off in outage probability between CC and V2V UEs. For a stringent CC outage constraint, V2V transmission is not permitted. As the CC constraint gets relaxed, the V2V routing method changes from IAR to SPR, and from the DL to the UL band. More specifically, the results show that for:

• CC outage constraint <5%: no V2V is permitted;

• CC outage constraint <12%: V2V using IAR in DL can achieve the lowest outage probability of 20%;

• CC outage constraint <15%: V2V using IAR in UL can achieve the lowest outage probability of 8%;

• CC outage constraint <40%: V2V using SPR in UL can achieve the lowest outage probability of 3%;

There is an intuitive trade-off in outage probability between CC and V2V, what has been improved is that by dynamically selecting the V2V routing method and transmission band, the V2V outage can be minimised. The V2V transmit band that causes the least

interference to CC is the DL band, but the V2V outage is reasonably high. As the outage constraint is relaxed in CC, there is a shift from interference aware transmit band and routing paths, to the shortest path in UL band.

8. Introduction to Mobile Network Security