The key motivations behind physical layer wireless security research

The key motivations behind physical layer security research, the role of physical layer security in future communication standards, and the relationship between physical layer security and upper layer security are summarized in the following points:

• - In all developed wireless communication systems and standards such as 2G, 3G, 4G, 5G, WiFi, WiMax, WiGig, etc., the main objectives of the physical layer transmission techniques and schemes have mainly been focused on achieving two key design requirements: 1) increasing data rates (higher capacity and spectral efficiency), and 2) enhancing reliability (lower error rates) along with reducing latency. These two key design requirements have been the primary driving factors for research and development in wireless communications up until recently. In fact, these requirements related to spectral efficiency and reliability are usually attained and met by using novel physical layer transmission techniques; while leaving security as an out of scope requirement that is left to be handled by upper layers. This conventional design paradigm has resulted in two phenomena: 1) what is called add-on security (i.e., security is an additional overhead added to different OSI layers) and 2) making security requirement a computer engineering issue rather than being a joint computer-communication issue as it must be. Consequently, the security services such as confidentiality and authentication have conventionally been achieved so far at the upper layers (such as application, transport, network, and MAC layers). For instance, to ensure the authenticity of a receiver, existing wireless systems typically employ multiple authentication approaches simultaneously at different layers, including MAC-layer authentication, network layer authentication, transport-layer authentication, and application layer authentication. A similar example applies to confidentiality, i.e., data protection from eavesdropping, where multiple confidentiality approaches are usually employed at different layers simultaneously to prevent data leakage to eavesdroppers. Particularly, at the Application/Presentation Layer, we have Secure SHell (SSH) where the encryption is in support of S-FTP, S-HTTP, PGP, S/MIME. At the Transport Layer, we have Secure Socket Layer (SSL) and Transport Layer Security (TLS). At the Network layer, we have IPSec Transport ESP and IPSec Tunnel ESP, which can be supported by the following encryption algorithms RC5, DES, AES, etc. At the Data Link Layer, we have Wire Equivalent Privacy (WEP), Temporal Key Integrity Protocol (TKIP), Counter-Mode MAC Protocol (CCMP). Obviously, this multi-layer security approach is costly and inefficient as it creates significant network bottlenecks, latency, signaling overhead, and increases computation complexity, especially for future wireless systems (i.e., beyond 5G) that are expected to provide highly secure, delay-sensitive, and low complexity applications and services (i.e., IoT-based services including URLLC and mMTC). To address these challenges, we propose to make security an inherent feature (not add-on) of the physical layer (not upper layers) transmission mechanism. This approach will motivate designing novel techniques that consider achieving the Quality of Service (QoS) requirements of different services at the lower physical layer in terms of not only reliability, capacity and latency, but also security.  Therefore, security, in this case, will be applied to the signals carrying the data bits, rather than being applied to the data bits themselves as is the case in upper layers cryptography approaches.  This would free the upper layers from any add-on security mechanism and make security not only an inherent feature of the transmission techniques at the physical layer but also complexity-independent where no matter what computational processing power the eavesdropper may have, the transmission techniques can still be secure.

 

The main performance metrics considered in the physical layer design of communication systems [Nokia]. Physical layer security is going to be the fourth critical dimension. 

In addition to the aforementioned points, we also have the following reasons that motivate the research work on physical layer security:

• - The key distribution and management processes for the legitimate parties in conventional encryption-based systems are extremely difficult and complex, especially in large-scale, dense, and heterogeneous wireless networks as is the case in future beyond 5G systems, where a massive number of smart devices are simultaneously connected to the network. This causes excessive complexity, high signaling overhead, and costly computational processes. Also, the management and control frames exchanged between communication entities are usually not very well protected.

• - Longer key length, which is usually preferable in cryptography approaches to increase the security strength, results in more waste of resources, apart from the fact that implementing security methods with Shannon’s perfect secrecy is extremely hard to be practically achieved with today’s huge data volume. For instance, to perfectly secure a message of 10 Gbyte in practice, we need to share and use a key of the same size and use it only once. When we transmit another message, we must generate another key and so on, which is costly and inefficient.

• -  The fast developments and advances in computing power devices reveal the fact that current secret key-based techniques can be cracked, no matter how much mathematically complex they are, especially when quantum computing becomes a reality. This would make all currently used encryption-based algorithms at risk and consequently all the applications that depend on using these algorithms for security.  

• - Cryptography-based security add extra delay and excessive computational power and complexity, making it inefficient and unsuitable to the IoT-based Tactile communication applications such as autonomous driving, remote surgery operation, controlling unmanned aerial vehicles (UAVs), etc. These future applications require the utmost secure communication with minimal latency. Particularly, given the extremely wide range of IoT-based wireless applications including industrial, medical, commercial, governmental, and military applications, designing practical security techniques is becoming an indispensable need for future xG systems.

• -  Besides, future mobile base stations (BS), as well as mobile devices and handsets (especially IoT devices), are expected to be noticeably different from the existing ones in terms of requirements, hardware capabilities, and channel nature. Thus, their security requirements and designs are also going to be significantly different.

All these issues together motivate the development and design of new practical security techniques at the lower physical (PHY) and MAC layers to protect and safeguard the wireless transmissions from future low-complexity devices such as IoT to BS (uplink) and from BS to IoT devices (downlink).

Author contact: jehad.hamamreh@gmail.com