An outlook on wireless coexistence with focus on medical devices

Abstract

The use of shared and unlicensed radio spectrum has been the impetus of innovative and widely used technologies (e.g., Wi-Fi, Bluetooth, and others)—driving the scene of ubiquitous connectivity for devices, sensors, and peripherals in an encompassing Internet of Things (IoT). However, coexisting technologies operating in unlicensed bands need to share limited spectrum resources while attempting to offer desired performance. This could prove challenging as channel access is not guaranteed, which raises concerns for wireless coexistence in sensitive applications like medical devices.

In this article, we provide a brief review of topics forming the current stage of wireless coexistence in research, industry, and regulatory circles. We present the standardization activities for the evaluation of wireless coexistence, and recent advancements of unlicensed spectrum technologies such as Wi-Fi, Bluetooth, and Long Term Evolution (LTE) in unlicensed bands, along with proposed evaluation methods.

2. Introduction

Ensuring medical device’s safety and effectiveness motivates the U.S. Food and Drug Administration (FDA) to engage technology trends and advances in standardization. Wireless communication continues to be an indispensable utility in modern life. Enabling medical devices to communicate wirelessly allows manufacturers to offer convenient and highly agile products to patients and care-givers. These efforts were significantly facilitated by the availability and affordability of wireless technologies that operate in unlicensed radio spectrum bands such as the 2.4 GHz and 5 GHz industrial, scientific, and medical (ISM) bands. The unlicensed nature of these bands alleviates the financial burden incurred by technologies that require exclusive access rights to radio channel resources (i.e., spectrum license) such as Long Term Evolution (LTE) cellular networks. However, regulations governing the use of unlicensed spectrum in the United States specify that a wireless device must accept any interference received, including interference that may cause undesired operation [1]. Accordingly, concerns for wireless coexistence could be traced to the lack of interference protection that forms the concept of unlicensed spectrum access, and the increasing popularity of medical device use of technologies operating in unlicensed bands (e.g., Wi-Fi, Bluetooth, etc.) Furthermore, the use of such technologies introduces potential hazards when associated with the medical use of a wireless function. These concerns are not adequately addressed by traditional electromagnetic immunity testing as the unwanted signals in a coexistence scenario originate from other wireless communication devices sharing the same spectrum [2]. Consequently, testing for wireless coexistence acknowledges the mutual interaction between coexisting wireless systems and how their respective implementations of a given wireless technology—across the open system interconnection (OSI) layers—mitigate the effects of sharing a transmission medium [3, 4]. The FDA is committed to promoting and protecting the public health and ensuring the safety and effectiveness of medical devices. Hence, the FDA issued a guidance document regarding the use of radio frequency wireless technology in medical devices, which recommends that device manufacturers address concerns of wireless coexistence in their premarket submissions [5]. The family of IEEE 802.11 standards (i.e., Wi-Fi) and Bluetooth are among the most popular unlicensed spectrum technologies used in medical devices such as pulse oximeters, monitors (electrocardiogram [ECG], blood pressure, etc.), pacemakers, infusion pumps, and others. Cellular communication systems like Long Term Evolution (LTE) are present in diverse use environments including healthcare. A recent trend in developments signals the interest of cellular technology designers in using unlicensed spectrum bands to extend their operation. Wireless medical devices operating in unlicensed spectrum like the 5 GHz band may need to coexist with users of emerging cellular technologies, whether these are medical devices or other equipment.

In this article, we provide a holistic picture of current and emerging topics of interest in the domain of wireless coexistence and emphasize medical devices as distinct users of unlicensed-spectrum technologies. The balance of this article is organized as follows. Section 3 addresses advances in technologies that are currently in wide-spread use and consensus standards for the evaluation of wireless coexistence. Emerging technologies that aim at exploiting unlicensed bands in addition to corresponding test methods that are recommended by technology-specific stakeholders are presented in Section 4. Section 5 concludes the paper.

3. Advances in established technologies and evaluation standards

To achieve desired performance, systems that use unlicensed spectrum for communication share spectrum resources through either alternating transmission times, using non-overlapping frequency channels, selecting a transmission power value that facilitates reception only in the local vicinity of the transmitter, or a combination of these and other techniques. Figure 1 illustrates how two nodes can share spectrum resources in time, frequency, and power/space. Reports in the literature detail a plethora of approaches to achieve spectrum sharing. The area of cognitive radio in TV whitespaces has particularly seen active research activities on the topic [6]. Examples of methods for spectrum sharing in the time dimension include the listen-before-talk (LBT) approach that Wi-Fi implements in the context of carrier sense multiple access with collision avoidance (CSMA/CA) for medium access control (MAC) protocol. In the frequency dimension, Bluetooth’s adaptive frequency hopping (AFH) avoids channels that suffer from interference and concentrate traffic on others with better conditions. Adaptive power control or simple physical separation between transmitting nodes could achieve coexistence in the power dimension. At the design stage, unlicensed spectrum technology developers attempt to systematically equip their products with coexistence-enabling features. In the IEEE 802 community—where technical development of Wi-Fi takes place—these efforts are organized and reviewed by IEEE 802.19 Wireless Coexistence Working Group. Technology development groups address coexistence with varying levels of input from stakeholders of other technologies sharing the same spectrum. Note that wireless coexistence in this article refers to the phenomena that occur when multiple systems share a given spectrum band as defined in [5]. In the following, we summarize the efforts and methodology of relevant general coexistence evaluation standards and highlight potential research directions given recent advances in Wi-Fi and Bluetooth.

Figure 1.

Spectrum resource sharing in the three fundamental physical dimensions of time, frequency, and power.

3.1. Evaluation standards

The evaluation of wireless coexistence for a medical device begins at the risk assessment stage to determine the hazards associated with implementing wireless functionality in the device. This process is described in detail in the technical information report TIR69 by the Association for the Advancement of Medical Instrumentation that was published in February 2017 [7]. Given the risk-based evaluation approach, AAMI TIR69 calls for more rigorous testing for high risk wireless functions. Wireless coexistence testing is performed according to the American National Standards Institute (ANSI) C63.27 standard for evaluation of wireless coexistence—published in May 2017. Notably, AAMI TIR69 primarily addresses medical devices while ANSI C63.27 has a general scope. Prior to testing, ANSI C63.27 calls for the definition of functional wireless performance “FWP” of the equipment-under-test, defined as the subset of the total functionality that uses the wireless capabilities of the equipment-under-test and would result in unacceptable consequences if degraded or disrupted [8]. In the case of medical devices, unacceptable consequences could include delay in—or prevention of—delivery of therapy, delay or failure to access medical device data, delay or failure in programming medical device behavior, and others. Key performance indicators (KPIs) are then established relevant to FWP and testing is performed by forming a shared communication channel (whether conducted using cables and RF components or radiated) between the communicating nodes of the system under test and those of an interfering system. The rigor of the test is controlled through parameters like the interfering system power level, traffic load (i.e., channel utilization), and the relative overlap between channels in frequency domain (e.g., one or more Wi-Fi channels overlapping several segments of Bluetooth channels). In [9], we provide a description of the test methods included in ANSI C63.27 and the risk assessment approach to coexistence testing according to AAMI TIR69. We note that in its current format ANSI C63.27 does not specifically address unlicensed extensions of LTE.

AAMI TIR69 and ANSI C63.27 help provide manufacturers, test labs, and the FDA with a common evaluation tool to assess wireless coexistence. Given the recent publication date of the two documents and the corresponding FDA’s recognition in August 2017, it is unclear yet how device manufactures are adopting this tool in their regulatory submissions.

3.2. Advances to established technologies

Recent advances in Wi-Fi and Bluetooth offer an expanding set of features to users while taking wireless coexistence into consideration. However, sophisticated capabilities could create novel challenges. For example, authors in [10] address IEEE 802.11ac hidden channel (HC) in Wi-Fi deployments with heterogeneous bandwidth allocations (e.g., one system is using 80 MHz channel while another is using 20 MHz channel) where asymmetric channel sensing could negatively affect some coexisting nodes. Similar and other features seem to be under consideration in the currently under-development IEEE 802.11ax that targets dense Wi-Fi deployments in the 2.4 and 5 GHz unlicensed bands [11]. Novel mechanisms—such as adaptively changing the clear channel assessment level of a node attempting to access the channel—call for novel coexistence testing techniques that are open for research to facilitate innovative applications in medical devices and the healthcare domain.

Bluetooth is a commonly used wireless technology in medical devices. Cost, low power consumption and dynamic interference avoidance are among the features promoting this popularity. Bluetooth operates by periodically changing its transmission channel (i.e., hopping), which helps circumvent collisions with other transmitters and facilitates coexistence. A review of relevant literature for Bluetooth coexistence is provided in [2]. A single Bluetooth channel can have a bandwidth of either 1 MHz or 2 MHz. The former is used by classic Bluetooth that hops every 625 μs on a set of 79 channels in the 2.4 GHz ISM band and the latter is used by Bluetooth Low Energy (BLE) that hops at a slower rate over 40 channels in the same band. The most recent version of the specifications developed by the Bluetooth Special Interest Group (SIG) is Bluetooth 5 that introduced more flexibility in controlling the tradeoff of increasing the data rate at the expense of coverage range. Mesh networking capability was also introduced as a universal solution to overcome limited coverage allowing the interconnection of equipment in large setups like a hospital that extends beyond the limits of a single link between two nodes. With the introduction of hardware implementations embedding Bluetooth 5, studies of experimental performance evaluations started to be reported in the literature [12]. However, research to assess wireless coexistence in mesh scenarios applicable to healthcare environments is yet to be addressed.

4. Emerging technologies and evaluation methods

The use of cellular communications in unlicensed spectrum bands was introduced by LTE-Unlicensed (LTE-U) and LTE Licensed Assisted Access (LTE LAA)—both targeting the unlicensed 5 GHz band. LTE-U and LAA are extensions to an existing LTE system to allow offloading of traffic to offer larger capacity to users at a low cost. The standardization of the 5^th generation (5G) cellular communication networks is ongoing by the 3^rd Generation Partnership Project (3GPP) to achieve unprecedented gains in throughput, latency, and number of users to meet the guidelines for 5G set forth by the International Telecommunication Union (ITU) [13]. 5G is shaping as a path towards unifying diverse cellular deployments, spectrum, services, and devices. Ground-breaking performance gains are promised to be delivered using a novel design for the radio air interface, namely New Radio (NR)—capable of running flexibly from sub-1 GHz large-coverage frequencies to the line-of-sight above-6 GHz millimeter waves. Central to NR is using unlicensed radio spectrum bands, where wireless operations are not guaranteed access and must accept interference from other users of the band. It is an open research question to report how this development unfolds relative to the previously introduced LTE LAA and how wireless coexistence will be designed and tested in the context of 5G NR. In this section, we summarize the coexistence evaluation methods proposed by the Wi-Fi alliance and the 3GPP.

4.1. Wi-Fi alliance LTE-U Coexistence Test Plan

LTE-U is an industry led proposal under the umbrella of the LTE-U Forum, established in 2014 by Verizon, Alcatel-Lucent, Ericsson, Qualcomm, and Samsung. LTE-U was designed to comply with LTE R10/11/12—commercially deployed at the time of development. Consequently, little modifications to LTE systems were sought, which include not using LBT approach before attempting channel access. LBT is not required by unlicensed bands regulations in the United States (see [1]). LTE-U attempts to achieve successful coexistence with Wi-Fi and other LTE-U systems using a duty-cycling method in which LTE-U monopolizes channel resources during ON periods. OFF periods are left for other coexisting systems to contend for resources. The algorithm that performs this operation is labeled Carrier Sense Adaptive Transmission (CSAT). During the OFF periods, CSAT monitors the wireless channel activities and uses observed traffic patterns to dynamically increase or decrease the ON period time.

To address the impact of an LTE-U system on a Wi-Fi network, the Wi-Fi alliance released a proposed test plan in 2016 [14]. The objective of the plan is to determine whether an LTE-U network impacts a Wi-Fi network any more than a Wi-Fi network impacts another Wi-Fi network. The plan recommends using a hybrid test setup that includes radiated over-the-air (OTA) and conducted signal path segments, as illustrated in Figure 2. This is similar in concept to the multiple chamber test method described in Annex C of ANSI C63.27. The communication flows over the licensed-spectrum part of the LTE-U system (i.e., Primary Cell [PCell]) to establish and maintain connectivity in addition to unlicensed communication (i.e., Secondary Cell [SCell]) that is evaluated for coexistence. Tests are designed to assess 1) the ability of the LTE-U system to select a vacant—or least utilized—communication channel, 2) prohibitive effects of LTE-U on establishing and exchanging data over a Wi-Fi connection, 3) dynamic adaptation of LTE-U channel utilization with varying Wi-Fi traffic, and 4) the impact of LTE-U on Wi-Fi throughput and latency sensitive voice traffic.

Figure 2.

LTE-U Coexistence evaluation plan proposed by the Wi-Fi alliance. Double sided rectangles in the figure represent semi-anechoic shielded boxes. STA and AP refer to Wi-Fi station and access point, eNB and UE refer to LTE-U basestation and user equipment.

4.2. 3GPP LAA coexistence tests

LAA was first detailed in the 3GPP Release 13 specifications and is intended for global deployment (i.e., it satisfies regulatory requirements globally) [15]. LAA operates in a secondary cell on the downlink (i.e., cellular basestation to user equipment) in the 5 GHz band, while signaling is maintained in the primary cell through licensed spectrum. LAA on the secondary cell offers best effort throughput and the primary cell offers guaranteed uplink quality of service (QoS) and reliable signaling. LAA implements LBT in a way similar to Wi-Fi: when a frame arrives in the transmission queue, the basestation senses the wireless channel to determine if it is busy or idle and only transmits after waiting for an initial period of time. In case the channel was busy, the basestation performs one or several back-off events until the channel becomes available.

3GPP Technical Report (TR) 36.789 presents multi-node tests relevant to LAA [16]. Similar to the RF conducted test method described in Annex B of ANSI C63.27, the test setup depicted in Figure 3 is based on establishing a conducted communication path between LAA nodes and nodes of a coexisting technology—primarily Wi-Fi—using coaxial cables, attenuators, and other RF circuitry. The communication path for the licensed-spectrum part of the setup is also conducted. A test run comprises two steps: 1) evaluate KPIs of two systems running the same technology (i.e., either Wi-Fi or LTE LAA), and 2) evaluate KPIs after replacing one of the systems by one that runs the other technology. Investigated KPIs include throughput, delay, jitter, and mean opinion score (MOS, to measure the quality of voice traffic).

Figure 3.

3GPP LAA multi-node coexistence testing plan

5. Summary

Wireless medical devices use technologies that operate in the 2.4 GHz and 5 GHz unlicensed bands. Ongoing advances are increasing the number of technology choices that a medical device could use or coexist with—thus expanding the scope of wireless coexistence evaluation to ensure medical device safety and effectiveness. This article has presented a brief overview of recent developments of established technologies in unlicensed bands such as Wi-Fi and Bluetooth, and the emerging use of unlicensed spectrum by cellular systems. Accordingly, proposed test methods by the Wi-Fi alliance and the 3GPP were summarized. We note the lack of unified methodology to assess wireless coexistence of emerging technologies. This might be addressed by the working group of ANSI C63.27—currently preparing for a second revision of the standard. Future research is needed to identify and mitigate potential effects of wireless coexistence on medical devices, thus enabling innovative applications in healthcare.