With the paradigm shift from Shannon’s legacy, semantic communication (SC) is emerging as one of the promising next-generation communication technologies. The new paradigm in communication technology allows the meaning of transmitted messages to be successfully delivered to a receiver. Hence, the semantic communication focuses on the successful delivery of transmitted messages such as human language communication. In order to realize such new communication, both transmitter and receiver should share the same background knowledge with each other. Recently, several researchers have developed task-specific SC systems by exploiting astonishing achievements in deep learning, which can allow the same knowledge to be shared between them. However, since such SC systems are specialized to handle specific applications, not all users can be serviced by the SC systems. Therefore, a network will face a coexistence of an SC system and a traditional communication (TC) system. In this paper, we investigate how introducing emerging SC systems affects the performance of the TC system from a network perspective. For analysis, we consider the signal-to-noise ratio (SNR) differently for the user served by an SC system and the user served by a TC system. Then, by using two different SNR equations, we formulate a max-min fairness problem in the coexistence of SC and TC systems. Via extensive numerical results, we compare the network performance of TC and SC users with and without SC systems, and then confirm that SC systems are indeed a promising next-generation communication alternative.

No abstract

No abstract

In a disaster site, terrestrial communication infrastructures are often destroyed or malfunctioning, and hence it is very difficult to detect the existence of survivors in the site. At such sites, UAVs are rapidly emerging as an alternative to mobile base stations to establish temporary infrastructure. In this paper, a novel deep-learning-based multi-source detection scheme is proposed for the scenario in which an UAV wants to estimate the number of survivors sending rescue signals within its coverage in a disaster site. For practicality, survivors are assumed to use off-the-shelf smartphones to send rescue signals, and hence the transmitted signals are orthogonal frequency division multiplexing (OFDM)-modulated. Since the line of sight between the UAV and survivors cannot be generally secured, the sensing performance of existing radar techniques significantly deteriorates. Furthermore, we discover that transmitted signals of survivors are unavoidably aysnchronized to each other, and thus existing frequency-domain multi-source classification approaches cannot work. To overcome the limitations of these existing technologies, we propose a lightweight deep-learning-based multi-source detection scheme by carefully designing neural network architecture, input and output signals, and a training method. Extensive numerical simulations show that the proposed scheme outperforms existing methods for various SNRs under the scenario where synchronous and asynchronous transmission is mixed in a received signal. For almost all cases, the precision and recall of the proposed scheme is nearly one, even when users’ signal-to-noise ratios (SNRs) are randomly changing within a certain range. The precision and recall are improved up to 100% compared to existing methods, confirming that the proposal overcomes the limitation of the existing works due to the asynchronicity. Moreover, for Intel(R) Core(TM) i7-6900K CPU, the processing time of our proposal for a case is 31.8 milliseconds. As a result, the proposed scheme provides a robust and reliable detection performance with fast processing time. This proposal can also be applied to any field that needs to detect the number of wireless signals in a scenario where synchronization between wireless signals is not guaranteed.