One of the 5G use cases, known as ultra-reliable communication (URC), is expected to support very low packet error rate on the order of 10 −5 with a 1 ms latency. In an industrial scenario, this would make possible replacing wired connections with wireless for controlling critical processes. Industrial environments with large metallic machinery and concrete structures can lead to deep shadowing and severe fading in the radio propagation channel, and thus pose a challenge for achieving the outage levels in connection with URC. In this paper, we present and analyze the large-scale propagation characteristics of two different industrial environments -open production space and dense factory clutter -based on measurements conducted at 2.3 and 5.7 GHz By including a large number of spatially distributed samples, as per our experimental approach, we show the importance of properly characterizing the large-scale fading outage for URC. For instance, we show that based on a simple one-slope distance dependent path loss model, the conventional log-normal model for large-scale shadow fading is by far too simple for this environment. Our results show that at the 10 −4 percentile, the tail of the shadow fading distribution can deviate by up to 10-20 dB from the log-normal model with respect to the average NLOS values (around 6 dB and 8 dB at 2.3 and 5.7 GHz, respectively). The simplicity of the one-slope path loss model, and its ability as we show, to express the trends with respect to scenarios, frequencies, and antenna heights, makes it an attractable option. However, there is a need for further experimental insight, possibly in combination with deterministic analysis, to get a better understanding of the large-scale fading for the study of URC in industrial environments.

The upcoming fifth-generation wireless technology application areas bring new communication performance requirements, mainly in terms of reliability and latency, but also in terms of radio planning, where the further detailed characterization of the wireless channel is needed. To address these demands, we developed an agile multi-node multi-antenna wireless channel sounding system, using multiple softwaredefined radio devices. The system consists of 12 testbed nodes which are controlled from a centralized testbed server. Each node features a control host computer and two multi-antenna universal software radio peripheral boards. By managing the transmission and reception of reference signals among all the distributed testbed nodes, the system can measure the channel conditions of all multiple independent radio links. At the same time, the distributed architecture of the testbed allows a large number of spatially distributed locations to be covered with only a few redeployments of the testbed nodes. As a consequence of this, the system favors the collection of a large number of distributed channel samples with limited effort within a short dedicated measurement time. In this paper, we detail the general testbed design considerations, along with the specific sounding signal processing implementations. As further support to the system design, we also include the results from different verification and calibration tests, as well as a real measurement application example. INDEX TERMS Channel sounding, multi-antenna, multi-node, SDR, USRP. I. INTRODUCTION In the last three decades, wireless communication systems have evolved from the 1 st generation to the 4 th generation, with the primary aim of improving user cellular broadband services. The upcoming 5 th generation (5G) systems are also expected to enhance the wireless connection capabilities towards connecting things. Wireless connected things in the context of 5G are envisioned to be employed in new application areas, e.g., smart factories, smart grids, and health-care, as prominent examples. Apart from new communication requirements in terms of latency and reliability, these new areas entail new and unconventional deployment scenarios [1]. These scenarios, including, for example, deployments in deep underground, inside factory clutter, at low antenna height or at different frequency bands, may be quite different from the typical urban/rural outdoor and indoor cases [2]. As a result, different propagation The associate editor coordinating the review of this manuscript and approving it for publication was Rui Wang.

In the last years, wireless communications in industrial scenarios are becoming an increasingly important market. Some of these communications have tight reliability requirements, but harsh propagation conditions in industrial scenarios represent a major challenge. In this paper, multi-connectivity is explored as a solution for assuring high reliability in industrial scenarios. Several multi-connectivity techniques are compared, using real channel measurements from two factories. Multiconnectivity comes at the cost of a reduced throughput in the mobile broadband services on the same network. In this paper, this impact is quantified to assess for the cost of implementing multi-connectivity.

No abstract

Network synchronization is a fundamental enabler of interference mitigation techniques that are required in ultradense deployment of small cells as targeted by upcoming 5th Generation (5G) wireless system. In this paper, we experimentally evaluate the possibility of acheiving in a distributed manner (i.e. without any external high precision reference clocks) a tight time alignment in a network of small cells. We are considering a software defined radio (SDR) network of 8 nodes, where each node adopts the Universal Software Radio Peripheral N200 (USRP N200) radio boards by Ettus Research, and the ASGARD software platform. Experimental results demonstrate the possibility of synchronizing the timing of the multiple nodes with a few μs precision, despite of the occurrencies of beacon losses.